evaluationofthequickenvironmentalexposureand...

Hindawi Publishing CorporationJournal of Environmental and Public HealthVolume 2012, Article ID 304314, 10 pagesdoi:10.1155/2012/304314

Research Article

Evaluation of the Quick Environmental Exposure andSensitivity Inventory in a Danish Population

Sine Skovbjerg,1 Nikolaj Drimer Berg,2 Jesper Elberling,1 and Karl Bang Christensen3

1 The Danish Research Centre for Chemical Sensitivities, Department of Dermato-Allergology, Gentofte Hospital,University of Copenhagen, 2820 Gentofte, Denmark

2 Research Centre for Prevention and Health, Glostrup Hospital, University of Copenhagen, 2600 Glostrup, Denmark3 Section of Biostatistics, Department of Public Health, University of Copenhagen, 1165 Copenhagen, Denmark

Correspondence should be addressed to Sine Skovbjerg, [email protected]

Received 28 April 2011; Revised 18 November 2011; Accepted 3 December 2011

Academic Editor: Robin Bernhoft

Copyright © 2012 Sine Skovbjerg et al. This is an open access article distributed under the Creative Commons Attribution License,which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Objectives. To evaluate a Danish translation of the Quick Environmental Exposure and Sensitivity Inventory (QEESI). Methods.The study included two groups: one comprised a random sample of 2000 individuals drawn from the Danish Civil RegistrationSystem; the other comprised 315 patients with chemical intolerance. Results. The evaluation suggested good reliability for the fourQEESI scales in terms of internal consistency and coefficients between test and retest scores. The discriminatory validity was thelargest for the Chemical (inhalant) Intolerance and Life Impact Scales. Using combined cut-off scores for these two scales provideda sensitivity of 92.1 and a specificity of 91.8 and yielded a prevalence of 8.2% in the population group. Conclusions. The Danishtranslation of the QEESI showed overall good reliability and validity. We recommend the use of the combined Chemical (inhalant)Intolerance and Life Impact Scales in future studies.

1. Introduction

Chemical intolerance, also referred to as multiple chemicalsensitivities (MCS), is a disorder characterized by reports ofnonspecific symptoms from various organ systems attributedby the individual to exposure to common airborne chem-icals [1]. In general the reported symptoms are attributedto previous chemical exposures and recur on subsequentexposures to the same or structurally unrelated chemicals atlevels normally considered to be nontoxic [1].

Symptoms of chemical intolerance are prevalent withestimates ranging from 9 to 33% in population-based stud-ies; however, such studies are few [2–7]. Physician-diagnosedMCS or reports of disabling consequences in the form ofsocial and occupational disruptions attributed to exposureto common airborne chemicals range from 0.5 to 6.3%[2–4, 7]. The reported symptoms typically vary betweenindividuals with women being more sensitive and reportingmore symptoms than do men [2, 5–7]. A typical symptompattern is thus difficult to establish. Nonspecific centralnervous system (CNS) complaints are frequently reported,

including fatigue, headache, and difficulty concentrating[2, 8, 9]. Other symptoms include pain and respiratorycomplaints [2, 5, 6]. An association between asthma andchemical intolerance has been reported in several studies [2,10, 11]. In a population-based twin study on the heritabilityof perfume-related respiratory symptoms, Elberling andcolleagues reported a heritability of 0.35 [12]. A mutualgenetic correlation of 0.39 was reported for perfume-relatedrespiratory symptoms and atopic dermatitis, suggestingsome genetic pleiotropy for these two factors. No geneticpleiotropy was found between perfume-related respiratorysymptoms, hand eczema, contact allergy, or asthma [12],suggesting that the association with asthma might be causedby mechanisms other than genetic susceptibility. Increasingevidence points to an association between MCS and symp-toms of psychological distress, that is, depressive symptoms,somatisation, negative affect, and anxiety [13–18], which arelikely to add to the level of overall functional disability.

The label “MCS” was initially proposed by Cullen basedon clinical observations [19]. Although more case definitionshave been proposed since the introduction of Cullen’s

2 Journal of Environmental and Public Health

criteria in 1987 [9, 20], none is currently widely accepted[20, 21]. The absence of widely accepted case criteria forestablishing the presence and degree of chemical intolerancechallenges epidemiological and clinical studies in this field.Several self-report questionnaires have been developed forresearch purposes [22–24]. The questionnaire that appears tohave been most widely applied is the Quick EnvironmentalExposure and Sensitivity Inventory (QEESI) developed byMiller and Prihoda [25, 26]. QEESI is a reliable andvalid self-administered questionnaire that was developed togauge the multisystem symptoms and multiple intolerancesoften reported in chemical intolerance [25, 26]. QEESIconsists of five scales measuring different domains related tochemical intolerance, that is, commonly reported symptoms,chemical (inhalant) intolerances, other intolerances (e.g.,allergies, foods, alcohol), life impact attributed to chemicalintolerances, and on-going exposures from routinely usedproducts (Masking Index). Four of the QEESI scales consistof ten items where responses are rated on an eleven pointscale ranging from “not at all a problem” (0) to “disablingsymptoms” [10], resulting in a score range from 0 to 100.The fifth, the Masking Index, also consists of ten items,but the response format is dichotomous (0 or 1), resultingin a score range from 0 to 10. QEESI has been translatedinto a number of different languages, that is, Swedish [27],Japanese [28, 29], and Spanish [30], of which the Swedishand Japanese versions have also been evaluated in terms ofvalidity and reliability. The Swedish study included a mildly(n = 67) and a moderately/severely chemically intolerantgroup (n = 126) and a control group (n = 90). Thestudy concluded that the Swedish version of QEESI is reliableand valid for investigating chemical intolerance [27]. TheJapanese study included a general population group (n =498) and an outpatient group with self-reported MCS (n =131) [28]. Based on principal components analyses, thisstudy concluded that three of the QEESI subscales, that is,Symptom Severity, Chemical (inhalant) Intolerances, andLife Impact, were valid. To the best of our knowledge, noother study has established normative data based on a largepopulation-based sample, and an evaluation of a Danishversion of QEESI will not only strengthen future studieson chemical intolerance but will also enable internationalcomparisons of data.

The objectives of the present study were (1) to evaluatea Danish translation of the QEESI in relation to validityand reliability, (2) to describe sensitivity and specificity, (3)to test whether asthma and high scores (based on Danishpopulation norms) on SCL-92 subscales of depression andsomatisation were associated with scores on QEESI, and (4)to establish normative data.

2. Methods

2.1. Participants. Two groups were invited to participatein the study: (1) individuals from the general populationand (2) individuals who had contacted the Danish ResearchCentre because of symptoms attributed to common airborne

chemicals, and patients with physician-diagnosed chemicalintolerance.

2.1.1. General Population. A random sample of 18–69-year-old (n = 2000) from the general population was drawn fromthe Danish Civil Registration system in January 2010.

2.1.2. Patients. The patient sample (n = 315) comprisedindividuals who had contacted the Danish Research Centrefor Chemical Sensitivities between January 1, 2006 andJanuary 1, 2010 (n = 183) because of reactions consistentwith chemical intolerance, and individuals who had receiveda diagnosis of chemical intolerance either at the CopenhagenUniversity Hospital, Rigshospitalet, or at Hamlet, PrivateHospital, Denmark, between January 1, 1990 and January 1,2009 (n = 132) by the same ear-nose-and throat specialist.

2.2. Measurements

2.2.1. A Danish Translation of QEESI. The original version ofQEESI [25, 26] was translated into Danish by a professionaltranslation agency. The Danish translation was subsequentlytailored to Danish usage and then translated back to theoriginal language by a different professional translator. Theback translation was then compared with the original versionof QEESI to identify potential sense-altering discrepancies.Finally, the Danish translation was pilot tested amongindividuals with chemical intolerance for comprehensionand ease of completion.

2.2.2. Symptom Checklist 92. Symptom Checklist 92 (SCL-92) subscales for depression and somatisation were included.These subscales comprise 25 items where responses are ratedon a 5-point Likert scale ranging from not at all to verymuch. The SCL-92 has been validated in a general Danishpopulation and normative data have been established [31,32].

2.2.3. Asthma. Questions on asthma were adopted fromthe Stage 1 questionnaire of the European CommunityRespiratory Health Study (ECRHS) [33]. Asthma wasdefined according to criteria employed by the ECRHS as anaffirmative answer to at least one of the following questions:(1) Have you been woken by an attack of shortness of breath atany time in the last 12 months? (2) Have you had an attack ofasthma in the last 12 months? (3) Are you currently taking anymedicine (including inhalers, aerosols, or tablets) for asthma?[34].

2.2.4. Procedure. A questionnaire was sent to the partici-pants on two occasions. The first test occasion included(1) the QEESI, (2) questions on socioeconomic position,categorized in accordance with the British Registrar General’sClassification I–V [35], (3) the SCL-92 somatisation anddepression subscales, and (4) the ECRHS asthma questions.Demographic data, for example, age and sex, were available.Two months after responding to the first questionnaire,a random sample of the respondents from the general

Journal of Environmental and Public Health 3

population (n = 200) and 140 patients who had respondedto the first questionnaire received a second questionnaire,which consisted of the QEESI only. The overall response rateto the first questionnaire was 64.5%. The response rates inthe population sample were 65.3% (n = 1305/2000) on thefirst test occasion and 61.0% (n = 122/200) on the secondtest occasion. The response rates in the patient sample were60.0% (n = 189/315) on the first test occasion and 80.0%(n = 112/140) on the second test occasion.

3. Statistical Analysis

3.1. Reliability and Validity. The internal consistency ofthe four QEESI scales (Chemical (inhalant) Intolerances,Symptom Severity, Other Intolerances, and Life Impact) wasevaluated using Cronbach’s alpha [36]. Coefficients were cal-culated for the patient sample and for age stratified samplesof the population. Test-retest reliability was evaluated usingPearson correlations.

The discriminatory validity of the QEESI was evaluatedusing bivariate logistic regression, and multivariate analyseswere also used for the Chemical (inhalant) Intolerance andLife Impact Scales. Criterion validity was addressed using thevariables asthma, somatisation, and depression, for whichassociations with chemical intolerance have been reported.These variables were dichotomized using the ECRHS asthmacriteria and the gender-based cut-off scores for the SCL-92 somatisation and depression subscales described byOlsen and colleagues [31, 32]. Further, cross-validation wasperformed by randomly dividing the data set in two andcomparing results with those obtained for the entire data set.

3.2. Differential Item Functioning. Differential item func-tioning (DIF) is the phenomenon that performance of itemsdiffers across subpopulations or that items measure differentthings for members of one subpopulation as opposed tomembers of another. Instruments containing such itemsmay have reduced validity for between-group comparisonsbecause scores may be indicative of attributes other thanthose the instrument is intended to measure [37]. We testedDIF by testing conditional independence given the totalscore. We used the partial gamma coefficient [38], suggestedby Kreiner when items are polytomous [39]. DIF withrespect to asthma was tested using the ECRHS criteria, anddepressive and somatising individuals were identified usingthe SCL-92 cut-off scores [31, 32].

Data were analysed using SPSS, version 15.0 for Windowsand SAS version 9.2.

4. Approval

The study was approved by the Danish Data ProtectionAgency. According to Danish legislation questionnaire stud-ies do not need approval from an ethics committee.

5. Results

5.1. Sample Characteristics. Characteristics of the patient-and population samples are shown in Table 1. Due to skeweddistributions, the medians for the five QEESI scales (theSymptom Severity, Chemical (inhalant) Intolerances, OtherIntolerances, Life Impact Scales, and the Masking Index) arepresented. Table 1 also includes sex and age distributionsfor the two samples, mean and median scores on thetwo SCL-92 subscales, as well as occupational social class(SES). Table 1 shows that scores on the QEESI and SCL-92 differed significantly in the expected direction betweenthe two samples. In analyses stratified by gender QEESIscores also differed significantly (P < 0.0001) between thetwo groups (data not shown). In terms of QEESI, scoresalso differed significantly (P < 0.0001) between women inthe population and patient samples as well as between men(data not shown). In regards to age, the patient sample wassignificantly older than the population and differences werealso seen in relation to SES classifications, which may be aconsequence of the differences seen in age.

5.2. Reliability. Cronbach alpha coefficients and medianscores on the four QEESI scales (Chemical (inhalant)Intolerances, Symptom Severity, Other Intolerances, andLife Impact) are shown in Table 2. The Cronbach alphacoefficients were overall high in both groups (range 0.64–0.94) for all four scales suggesting good internal consistency(Table 2).

Pearson correlation analyses of test-retest reliabilityshowed statistically significant coefficients for the five scales:the Chemical (inhalant) Intolerances Scale (0.94, n = 230),the Symptom Severity Scale (0.89, n = 234), the OtherIntolerances Scale (0.89, n = 233), the Life Impact Scale(0.96, n = 232), and the Masking Index (0.84, n = 234).

5.3. Validity. The discriminatory validity of the five QEESIscales is shown in Table 3. In the simple logistic regressionanalyses, the discriminatory power was the largest for theChemical (inhalant) Intolerance and the Life Impact Scale,and these two scales were therefore selected for subsequentmultivariate analyses. Calculating other pairwise compar-isons resulted in lower values than the one specified for theChemical (inhalant) Intolerance and the Life Impact Scales(data not shown). Including more scales in the analysis didnot substantially change the result since the maximum valueobtained for all five scales was 0.98 (data not shown).

To test whether other variables found to be associatedwith chemical intolerance, that is, asthma, somatisation, anddepression, would influence the discriminatory validity ofthe Chemical (inhalant) Intolerance and the Life ImpactScales, other subsequent statistical analyses were performed.Area under the ROC curve was calculated using the ECRHSasthma criteria and the gender-based cut-off scores for theSCL-92 somatisation and depression subscales, as describedby Olsen and colleagues [31, 32], in the analyses. Thefollowing results were obtained: for the Chemical (inhalant)Intolerance Scale the area under the ROC curve for asthma


Table 1: Characteristics of the patient and the population samples.

Population sample Patient sampleP-value1

Men Women Total Men Women Total

600 705 1305 25 163 188

Age, mean (sd) 47.4 (14.1) 46.8 (14.3) 47.1 (14.2) 52.4 (14.7) 56.0 (10.8) 55.5 (11.5) <0.0001

QEESI (median) P-value2

Symptoms 9.0 14.0 11.0 35.0 48.0 47.0 <0.0001

Chemical Int. 11.0 15.0 13.0 81.6 82.2 82.1 <0.0001

Other Int. 6.0 12.0 10.0 27.3 35.5 35.0 <0.0001

Life Impact 0.0 3.0 2.0 70.0 64.0 65.0 <0.0001

Masking Index 4.0 4.0 4.0 2.0 2.0 2.0 <0.0001

SCL-92 (median)

Depression 0.15 0.31 0.23 0.62 0.69 0.69 <0.0001

Somatisation 0.25 0.33 0.33 0.83 1.0 1.0 <0.0001

SCL-92 (mean)

Depression 0.39 0.53 0.47 1.0 0.86 0.88 —

Somatisation 0.38 0.51 0.45 1.0 1.13 1.12 —

Occupational social class (n (%)) P-value3

<0.0001

I + II: 224 (17.2) 17 (9.0)

III + IV: 426 (32.6) 35 (18.5)

V + VI + VII: 566 (43.4) 124 (65.6)

Missing: 89 (6.8) 13 (6.9)

Occupational social class: I + II: professionals and executives and medium-level white-collar employees; III + IV: low-level white-collar employees and skilledworkers; V + VI + VII: unskilled and semiskilled workers, individuals receiving pension or disability benefits, and students.1Independent samples t-test for equality of means (total) between population and patient samples.2Mann-Whitney test (total) comparing population and patient samples.3Chi-squared test comparing population and patient samples.

Table 2: Median scores and scale reliability coefficients (Cronbach’s alpha).

Scale group NSymptom scale Chemical intolerance scale Other intolerance scale Life Impact scale

Median (IQR)∗∗ Alpha∗∗∗ Median (IQR) Alpha Median (IQR) Alpha Median (IQR) Alpha

Patient sample 189 47 (30–64) 0.84 80 (62–91) 0.91 34 (20–53) 0.83 64 (45–80) 0.89

Population 1309 11 (5–23) 0.86 13 (4–30) 0.92 10 (3–19) 0.77 2 (0–8) 0.86

Population∗

−30 201 12 (5–23) 0.83 11 (4–24) 0.87 11 (3–19) 0.64 4 (0–10) 0.74

30–40 218 11 (4–21) 0.85 15 (5–31) 0.93 11 (5–22) 0.79 2 (0–8) 0.86

40–50 288 11 (6–22) 0.89 14 (5–30) 0.92 11 (5–18) 0.79 2 (0–8) 0.90

50–60 312 12 (5–23) 0.86 14 (4–32) 0.94 8 (3–17) 0.78 1 (0–9) 0.87

60− 290 10 (4–23) 0.86 12 (1–29) 0.93 5 (0–17) 0.75 0 (0–5) 0.81∗

Population sample grouped by age.∗∗Interquartile range.∗∗∗Cronbach’s alpha.

was 0.93 (95% CI 0.90–0.95), for somatisation 0.89 (95% CI0.86–0.94) (women) and 0.91 (95% CI 0.88–0.94) (men),and for depression 0.94 (95% CI 0.91–0.97) (women) and0.94 (95% CI 0.91–0.96) (men); for the Life Impact Scale thearea under the ROC curve for asthma was 0.94 (95% CI 0.91–0.96), for somatisation 0.88 (95% CI 0.84–0.93) (women)and 0.91 (95% CI 0.88–0.94) (men), and for depression

0.92 (95% CI 0.88–0.95) (women) and 0.93 (95% CI 0.89–0.95) (men) (data not shown). These results suggest thatthe area under the ROC curve is slightly lowered withcoexisting asthma, depression, or somatisation; neverthelessthe discriminatory validity of QEESI is still good. Randomlydividing the data set in two yielded results that correspondedwith the results obtained for the entire data set: Chemical


Table 3: Discriminatory power of the five QEESI scales either when used alone (univariate analyses) or when combined in a multivariatelogistic regression model.

Scale univariate P-value Odds ratio (95% CI) one-point increase Area under ROC curve

Symptom severity <0.0001 1.07 (1.06–1.08) 0.88 (0.85–0.90)

Chemical intolerances <0.0001 1.11 (1.09–1.12) 0.97 (0.95–0.98)

Other Intolerances <0.0001 1.07 (1.06–1.08) 0.84 (0.81–0.87)

Life Impact <0.0001 1.10 (1.09–1.12) 0.97 (0.96–0.98)

Masking Index (rev)∗ <0.0001 2.48 (2.16–2.06) 0.81 (0.78–0.84)

Multivariate

Chemical intolerances <0.0001 1.06 (1.05–1.08) 0.98

Life Impact <0.0001 1.06 (1.05–1.07)∗

Scores on the Masking Index were reversed in the statistical analyses.

Table 4: Differential item functioning (DIF). Only significant results are shown.

Scale ItemAsthmatics Depressives Somatisers

partial γ coefficient

Chemical int.item 2 (tobacco smoke) 0.17 (se = 0.06) −0.16 (se = 0.06) —

item 4 (gasoline) −0.23 (se = 0.06) — —

item 8 (tar) — 0.16 (se = 0.04) −0.17 (se = 0.06)

Life Impact

item 2 (work ability) — 0.14 (se = 0.06) —

item 4 (choice of clothing) — −0.31 (se = 0.12) —

item 6 (choice of products) −0.16 (se = 0.08) — —

item 8 (choice of hobbies) 0.20 (se = 0.09) — —

item 9 (relation with spouse) — — −0.19 (se = 0.10)

Symptom severity

item 1 (muscle and joint pain) 0.23 (se = 0.06) −0.19 (se = 0.06) —

item 2 (mucosal or respiratory) 0.47 (se = 0.04) −0.30 (se = 0.07) —

item 4 (stomach and digestive) — −0.17 (se = 0.07) —

item 5 (concentration/memory) −0.16 (se = 0.06) 0.36 (se = 0.06) —

item 6 (tension and nervousness) −0.15 (se = 0.07) 0.79 (se = 0.04) —

item 7 (balance or coordination) — 0.21 (se = 0.07) 0.24 (se = 0.08)

item 10 (genital and urinary) −0.20 (se = 0.06) — −0.27 (se = 0.08)

Other Int.

item 3 (unusual cravings) −0.33 (se = 0.06) — —

item 4 (feeling ill after meals) — 0.42 (se = 0.08) 0.39 (se = 0.07)

item 6 (feeling ill) −0.21 (se = 0.08) 0.24 (se = 0.09) —

item 7 (alcoholic drinks) — 0.27 (se = 0.09) —

item 10 (allergic reactions) 0.36 (se = 0.06) −0.21 (se = 0.10) —

Masking Indexitem 2 (alcoholic intake) −0.17 (se = 0.08) −0.54 (se = 0.09) —

item 4 (fragranced products) −0.51 (se = 0.06) −0.35 (se = 0.11) —

item 10 (routine use of medicine) 0.62 (se = 0.05). 0.72 (se = 0.06) —

(inhalant) Intolerance Scale (OR 1.07, 95% CI 1.05–1.09)and the Life Impact Scale (OR 1.05, 95% CI 1.03–1.07) (Areaunder the ROC curve 0.98).

Construct validity was tested by analysing differentialitem functioning (DIF), which investigates if item scores areaffected by external variables. DIF was tested in asthmaticsand in depressives and somatisers using the SCL-92 cut-off scores for caseness [31, 32]. Only statistically significantresults are presented in Table 4.

5.4. Sensitivity and Specificity. Figure 1 shows the distribu-tion of scores in the two groups and cut-off values for allfour scales. Using all scales provided a sensitivity of 92.1%and a specificity of 93.1%. The ROC curves for the Chemical(inhalant) Intolerance and Life Impact Scales are shown inFigure 2. The 95% sensitivity and specificity and optimal cut-off scores for the Chemical (inhalant) Intolerance and theLife Impact Scales when used separately or when combinedare shown in Table 5. When used separately, the cut-off


5

10

15

20

30

25

0

5

10

15

20

30

25

0

0 10 20 30 40 50 60 70 80 90 100

Cut-point

Cut-point

Pati

ent

sam

ple

(%)

Gen

eral

pop

ula

tion

(%

)

(a) Chemical (inhalant) Intolerance Scale

0 10 20 30 40 50 60 70 80 90 100

Cut-point

Cut-point

0

10

20

30

40

50

60

70

0

10

20

30

40

50

60

70

Pati

ent

sam

ple

(%)

Gen

eral

pop

ula

tion

(%

)

(b) Life Impact Scale

Cut-point

Cut-point

5

10

15

20

30

35

25

0

5

10

15

20

35

30

25

0

0 10 20 30 40 50 60 70 80 90 100

Pati

ent

sam

ple

(%)

Gen

eral

pop

ula

tion

(%

)

(c) Symptom Severity Scale

Cut-point

Cut-point

5

10

15

20

35

30

40

25

0

5

10

15

20

35

30

40

25

0

0 10 20 30 40 50 60 70 80 90 100

Pati

ent

sam

ple

(%)

Gen

eral

pop

ula

tion

(%

)

(d) Other Intolerances Scale

Figure 1: Distribution of the two study samples responses to the four QEESI scales.


0

0.2

0.4

0.6

0.8

1

0 0.2 0.4 0.6 0.8 1

Sen

siti

vity

1-specificity

Life impact

(a)

0

0.2

0.4

0.6

0.8

1

0 0.2 0.4 0.6 0.8 1

Sen

siti

vity

1-specificity

Chemical intolerance

(b)

Figure 2: ROC curves for the Chemical (inhalant) Intolerance andLife Impact Scales.

values that provided the highest sensitivity and specificityfor the two scales were 47 and 21 respectively (Table 5).Combining the two scale scores by using cut-offs of 35(Chemical Intolerance scale) and 14 (Life Impact Scale)provided highest sensitivity and a specificity (Table 5). Millerand Prihoda [25, 26] used logistic regression to estimate aweighted sum of QEESI scales and an interaction term (theproduct of two scales) that could be used to provide anoptimal cut-point. We used logistic regression finding nosignificant interactions. In our data this weighted approachyields a sensitivity of 94% and specificity of 91% by firstcomputing R = −3.9665 + 0.0619∗ chemicalintolerance −0.0342∗ otherintolerance + 0.6104∗ maskingindex + 0.0767∗

lifeimpactscale − 0.00242∗ symptoms and then computingthe predicted probability prpr = exp(R)/(1 + exp(R)) and

Table 5: Sensitivity, specificity, and optimal cut-off values for thechemical intolerance scale and the Life Impact Scale.

Chemical intolerance cut-offscores

Sensitivity (%) Specificity (%)

37 95.2 82.8

47 89.3 89.4

58 83.9 95.2

Life Impact Scale cut-offscores

Sensitivity Specificity

14 95.8 86.2

21 91.0 90.9

31 86.8 95.2

Combined scale scores Sensitivity Specificity

Chemical intolerance cut-off35/Life Impact Scale cut-off14

92.1 91.8

classifying a subject as “chemically sensitive” if prpr > 0.09.These analyses suggest that the difference between ourapproach and Miller and Prihoda’s is minimal.

6. Discussion

The evaluation of the Danish version of QEESI suggestedgood reliability for the four scales, that is, Chemical(inhalant) Intolerances, Life Impact, Symptom Severity, andOther Intolerances, in terms of internal consistency andcoefficients between test and retest scores.

The overall response rate to the first questionnaire was64.5%. For the sample characteristics, the patients weresignificantly older than the general population sample anddifferences were also found in relation to SES, which may be aconsequence of the age differences. In accordance with resultsreported in other studies, the patient group also scoredsignificantly higher on the SCL-92 subscales [18, 40]. Scoreson the QEESI differed between the samples in the expecteddirection as the patients scored significantly higher on allfour scales, that is, the Chemical (Inhalant) Intolerances, LifeImpact, Symptom Severity and Other Intolerances Scales,whereas the population scored significantly higher on theMasking Index.

Our results on the Cronbach alpha coefficients for allfour scales and test-retest reliability showed good internalconsistency and reliability and correspond to the resultsobtained in other studies evaluating the QEESI. The Cron-bach alphas obtained in this study ranged from 0.64 to 0.94in the population sample with a tendency to lower scoresin the youngest age group, whereas the range in the patientsample was 0.83 to 0.91. Miller and Prihoda reported acorresponding range of 0.89–0.97 for the original Americanversion of the questionnaire [25, 26]. Evaluating a Swedishversion of the QEESI, Nordin and Andersson reported arange of 0.74 to 0.95 [27], and in the Japanese versionthe range was 0.87 to 0.94 for the Chemical (inhalant)Intolerances, the Life Impact, and the Symptom Severityscales [28].


The discriminatory validity was the largest for the Chem-ical (inhalant) Intolerance and Life Impact Scales. Testingthe influence of other variables, that is, asthma, depression,and somatisation, by calculating area under the ROC curvedid not substantially change the results. Using combinedcut-off scores of 35 for the Chemical (inhalant) IntoleranceScale and 14 for the Life Impact Scale provided the bestsimultaneous sensitivity and specificity, that is, 92.1 and 91.8,respectively. The corresponding sensitivity and specificityfor all five QEESI scales were 92.1% and 93.1%. Millerand Prihoda found that the discriminatory power for theSymptom Severity and the Chemical (inhalant) IntoleranceScales was largest [26]. They reported a sensitivity of 83.2%and a specificity of 84.2% using a cut-off score of≥40 for theChemical (inhalant) Intolerance Scale [26]. Using the cut-off scores collectively for the Chemical (inhalant) IntoleranceScale (≥40), the Symptom Severity Scale (≥40) and theOther Intolerance Scale (≥25) provided a sensitivity of 67.2%and a specificity of 90.9% [26]. In the Japanese evaluation ofthree of the QEESI scales, Hojo et al. reported the highestdiscriminatory ability for the Symptom Severity Scale with acut-off score of ≥20, which provided a sensitivity of 84.8%and a specificity of 84.0% [29]. Sensitivity and specificity forthe Life Impact Scale were 84.8% and 85.7%, respectively,with a cut-off score of ≥10. Contrary to our findings andthe findings by Miller and Prihoda, the Japanese versionof the Chemical (inhalant) Intolerance Scale had a lowsensitivity (73.4%) and specificity (69.6) using a cut-off scoreof ≥40 [29]. The cut-off scores applied in the Japanese studywere defined uniquely for the Japanese translation. Nordinand Andersson reported good discriminatory power for theSymptom Severity, Chemical (inhalant) Intolerance, and LifeImpact Scales [27]. The different findings may reflect cross-cultural differences in the responses to QEESI or, perhapsmore likely, reflect differences in study populations inrelation to the selection and definition of cases. Neverthelesswhen applying the QEESI in epidemiological studies or inclinical research, our results suggest that using the combinedcut-offs scores for the Chemical (inhalant) Intolerance andthe Life Impact Scales provides a shorter and equally strongalternative.

Construct validity was tested by analysing differentialitem function (DIF). Item function is supposed to beinvariant of other, and in this regard, irrelevant constructs[41]. Our analyses suggested that scores on several items onall five scales were influenced if the respondent had asthmaaccording to the ECRHS criteria or had scores above thecut-off values for caseness on the SCL-92 subscales, whichmay have a negative impact on the construct validity ofthe Danish translation of QEESI. The use of the ECRHSdefinition on asthma has been validated with bronchialhyperresponsiveness to methacholine (BHR) [33] but notvalidated among individuals with chemical intolerance andmight therefore overestimate a correlation. However, positivecorrelations between asthma and chemical intolerance havebeen described in studies using other self-reported asthmadefinitions [7, 42] as well as objective measurements of BHR[43]. Using the standards for interpretation of DIF applied byBjorner et al. [41], the magnitude of DIF for the three items

on the Chemical (inhalant) Intolerance Scale, that showedindication of DIF, was none or negligible. The same appliedfor the Life Impact Scale except for one item (choice ofclothing) in relation to depression, which was in the slightto moderate range. However, taken together the magnitudeof DIF appears to be of little importance for the constructvalidity of these two scales. For the remaining three scalesthe sizes of the gamma coefficients suggested that DIF maybe a problem. However, this study is the first to test DIFin relation to the QEESI. Therefore, we cannot compareour results with others and thereby determine whether DIFoccurs in other translations of the questionnaire than theDanish version. Accordingly, we recommend that futurestudies on QEESI address this issue. Altogether our resultsprovide additional evidence of the reliability and validityof QEESI as a clinical survey tool for MCS. The size ofthe study and the response rate to the questionnaire onboth the first and the second test occasion support thevalidity of our results. However, like most questionnaire-based studies, the information gathered relies upon self-reported and retrospective data, which must be kept in mindwhen interpreting the results. While reliability and validityof the different translations of QEESI have proven to be goodand thereby support the use of the questionnaire in futurestudies, differences in the case definitions applied in thestudies still point to difficulties in the comparisons of resultsacross countries. As stated by Miller and Prihoda in theirstudy published in 1999, the lack of a uniform approach foridentifying individuals with chemical intolerance is a barrierfor progress in this area [26]. Thus more research in thisarea is needed to establish internationally agreed diagnosticcriteria. Meanwhile, the QESSI provides a good research toolwith a response format that allows for continuous scores thatmay also be used in the evaluation of the effectiveness oftreatments.

In conclusion, the Danish translation of the QEESIshowed overall good reliability and validity, which is inaccordance with the results reported in other studies. Ouranalyses of construct validity suggested that there may beproblems with DIF in three of the QEESI scales. As our studyis the first to conduct these analyses, we cannot concludewhether this applies only to the Danish translation. Wetherefore recommend that future studies on QEESI addressthis issue. For research purposes, we recommend use ofthe combined Chemical (inhalant) Intolerance and the LifeImpact Scales scores, which provided a sensitivity of 92.1 anda specificity of 91.8 in this study.

Abbreviation List

MCS: Multiple chemical sensitivityQEESI: Quick Environmental Exposure and

Sensitivity InventorySCL-92: Symptom Checklist 92.

Acknowledgment

The authors are grateful to Anne Marie Topp for her invalu-able assistance in the data collection process and to Søren


Vesterhauge, ENT consultant, D.M.S., for his assistance inenrolling MCS patients in the study.

References

[1] R. A. Graveling, A. Pilkington, J. P. K. George, M. P. Butler, andS. N. Tannahill, “A review of multiple chemical sensitivity,”Occupational and Environmental Medicine, vol. 56, no. 2, pp.73–85, 1999.

[2] N. D. Berg, A. Linneberg, A. Dirksen, and J. Elberling,“Prevalence of self-reported symptoms and consequencesrelated to inhalation of airborne chemicals in a Danishgeneral population,” International Archives of Occupationaland Environmental Health, vol. 81, no. 7, pp. 881–887, 2008.

[3] S. M. Caress and A. C. Steinemann, “A national populationstudy of the prevalence of multiple chemical sensitivity,”Archives of Environmental Health, vol. 59, no. 6, pp. 300–305,2004.

[4] S. M. Caress and A. C. Steinemann, “Prevalence of multiplechemical sensitivities: a population-based study in the South-eastern United States,” American Journal of Public Health, vol.94, no. 5, pp. 746–747, 2004.

[5] C. Hausteiner, S. Bornschein, J. Hansen, T. Zilker, and H.Forstl, “Self-reported chemical sensitivity in Germany: apopulation-based survey,” International Journal of Hygiene andEnvironmental Health, vol. 208, no. 4, pp. 271–278, 2005.

[6] A. Johansson, A. Bramerson, E. Millqvist, S. Nordin, and M.Bende, “Prevalence and risk factors for self-reported odourintolerance: the Skovde population-based study,” InternationalArchives of Occupational and Environmental Health, vol. 78, no.7, pp. 559–564, 2005.

[7] R. Kreutzer, R. R. Neutra, and N. Lashuay, “Prevalence ofpeople reporting sensitivities to chemicals in a population-based survey,” American Journal of Epidemiology, vol. 150, no.1, pp. 1–12, 1999.

[8] N. D. Berg, A. Linneberg, A. Dirksen, and J. Elberling,“Phenotypes of individuals affected by airborne chemicals inthe general population,” International Archives of Occupationaland Environmental Health, vol. 82, no. 4, pp. 509–517, 2009.

[9] M. Lacour, T. Zunder, K. Schmidtke, P. Vaith, and C. Scheidt,“Multiple chemical sensitivity syndrome (MCS)—suggestionsfor an extension of the US MCS-case definition,” InternationalJournal of Hygiene and Environmental Health, vol. 208, no. 3,pp. 141–151, 2005.

[10] S. M. Caress and A. C. Steinemann, “Asthma and chemicalhypersensitivity: prevalence, etiology, and age of onset,”Toxicology and Industrial Health, vol. 25, no. 1, pp. 71–78,2009.

[11] J. Elberling, A. Linneberg, H. Mosbech et al., “Airbornechemicals cause respiratory symptoms in individuals withcontact allergy,” Contact Dermatitis, vol. 52, no. 2, pp. 65–72,2005.

[12] J. Elberling, A. Lerbaek, K. O. Kyvik, and J. Hjelmborg, “A twinstudy of perfume-related respiratory symptoms,” InternationalJournal of Hygiene and Environmental Health, vol. 212, no. 6,pp. 670–678, 2009.

[13] J. Bailer, M. Witthoft, and F. Rist, “Psychological predictorsof short- and medium term outcome in individuals withidiopathic environmental intolerance (IEI) and individualswith somatoform disorders,” Journal of Toxicology and Envi-ronmental Health Part A, vol. 71, no. 11-12, pp. 766–775, 2008.

[14] J. Bailer, M. Witthoft, C. Bayerl, and F. Rist, “Syndromestability and psychological predictors of symptom severity

in idiopathic environmental intolerance and somatoformdisorders,” Psychological Medicine, vol. 37, no. 2, pp. 271–281,2007.

[15] S. Skovbjerg, R. Zachariae, A. Rasmussen, J. D. Johansen, andJesper Elberling, “Attention to bodily sensations and symptomperception in individuals with idiopathic environmental intol-erance,” Environmental Health and Preventive Medicine, vol.15, no. 3, pp. 141–150, 2010.

[16] E. C. V. Vliet, K. Kelly-McNeil, B. Natelson, H. Kipen, andN. Fiedler, “Anxiety sensitivity and depression in multiplechemical sensitivities and asthma,” Journal of Occupational andEnvironmental Medicine, vol. 44, no. 10, pp. 890–901, 2002.

[17] K. Osterberg, R. Persson, B. Karlson, F. Carlsson Eek, andP. Ørbæk, “Personality, mental distress, and subjective healthcomplaints among persons with environmental annoyance,”Human and Experimental Toxicology, vol. 26, no. 3, pp. 231–241, 2007.

[18] I. R. Bell, J. M. Peterson, and G. E. Schwartz, “Medical historiesand psychological profiles of middle-aged women with andwithout self-reported illness from environmental chemicals,”Journal of Clinical Psychiatry, vol. 56, no. 4, pp. 151–160, 1995.

[19] M. R. Cullen, “The worker with multiple chemical sensitiv-ities: an overview,” Occupational Medicine, vol. 2, no. 4, pp.655–661, 1987.

[20] L. Bartha, W. Baumzweiger, D. S. Buscher et al., “Multiplechemical sensitivity: a 1999 consensus,” Archives of Environ-mental Health, vol. 54, no. 3, pp. 147–149, 1999.

[21] R. Kreutzer, “Idiopathic environmental intolerance: case defi-nition issues,” Occupational Medicine, vol. 15, no. 3, pp. 511–517, 2000.

[22] S. Nordin, E. Millqvist, O. Lowhagen, and M. Bende, “Ashort chemical sensitivity scale assessment of airway sensoryhyperreactivity,” International Archives of Occupational andEnvironmental Health, vol. 77, no. 4, pp. 249–254, 2004.

[23] M. J. E. Andersson, L. Andersson, M. Bende, E. Millqvist, andS. Nordin, “The idiopathic environmental intolerance symp-tom inventory: development, evaluation, and application,”Journal of Occupational and Environmental Medicine, vol. 51,no. 7, pp. 838–847, 2009.

[24] J. Bailer, M. Witthoft, and F. Rist, “The chemical odorsensitivity scale: reliability and validity of a screening instru-ment for idiopathic environmental intolerance,” Journal ofPsychosomatic Research, vol. 61, no. 1, pp. 71–79, 2006.

[25] C. S. Miller and T. J. Prihoda, “A controlled comparisonof symptoms and chemical intolerances reported by GulfWar veterans, implant recipients and persons with multiplechemical sensitivity,” Toxicology and Industrial Health, vol. 15,no. 3-4, pp. 386–397, 1999.

[26] C. S. Miller and T. J. Prihoda, “The environmental exposureand sensitivity inventory (EESI): a standardized approachfor measuring chemical intolerances for research and clinicalapplications,” Toxicology and Industrial Health, vol. 15, no. 3-4,pp. 370–385, 1999.

[27] S. Nordin and L. Andersson, “Evaluation of a Swedishversion of the quick environmental exposure and sensitivityinventory,” International Archives of Occupational and Environ-mental Health, vol. 83, no. 1, pp. 95–104, 2010.

[28] S. Hojo, H. Kumano, H. Yoshino, K. Kakuta, and S. Ishikawa,“Application of quick environment exposure sensitivity inven-tory (QEESI) for Japanese population: study of reliability andvalidity of the questionnaire,” Toxicology and Industrial Health,vol. 19, no. 2-6, pp. 41–49, 2003.

[29] S. Hojo, K. Sakabe, S. Ishikawa, M. Miyata, and H. Kumano,“Evaluation of subjective symptoms of Japanese patients with


multiple chemical sensitivity using QEESI,” EnvironmentalHealth and Preventive Medicine, vol. 14, no. 5, pp. 267–275,2009.

[30] S. Nogue, J. Fernandez-Sola, E. Rovira, E. Montori, J. M.Fernandez-Huerta, and P. Munne, “Multiple chemical sensi-tivity: study of 52 cases,” Medicina Clinica, vol. 129, no. 3, pp.96–98, 2007.

[31] L. R. Olsen, E. L. Mortensen, and P. Bech, “The SCL-90 andSCL-90R versions validated by item response models in aDanish community sample,” Acta Psychiatrica Scandinavica,vol. 110, no. 3, pp. 225–229, 2004.

[32] L. R. Olsen, E. L. Mortensen, and P. Bech, “Mental distress inthe Danish general population,” Acta Psychiatrica Scandinav-ica, vol. 113, no. 6, pp. 477–484, 2006.

[33] J. Pekkanen, J. Sunyer, J. M. Anto et al., “European communityrespiratory health study operational definitions of asthma instudies on its aetiology,” European Respiratory Journal, vol. 26,no. 1, pp. 28–35, 2005.

[34] J. Sunyer, D. Jarvis, J. Pekkanen et al., “Geographic variationsin the effect of atopy on asthma in the European CommunityRespiratory Health Study,” Journal of Allergy and ClinicalImmunology, vol. 114, no. 5, pp. 1033–1039, 2004.

[35] U. Christensen, L. Schmidt, C. Ørsted Hougaard, M. Krieg-baum, and B. E. Holstein, “Socioeconomic position andvariations in coping strategies in musculoskeletal pain: a cross-sectional study of 1287 40- and 50-year-old men and women,”Journal of Rehabilitation Medicine, vol. 38, no. 5, pp. 316–321,2006.

[36] L. J. Cronbach and W. G. Warrington, “Time-limit tests:estimating their reliability and degree of speeding,” Psychome-trika, vol. 16, no. 2, pp. 167–188, 1951.

[37] P. W. Holland and H. Wainer, Differential Item Functioning,Lawrance Erlbaum Associates, London, UK, 1993.

[38] J. A. Davis, “A partial coefficient for Goodman and Kruskal’sgamma,” Journal of the American Statistical Association, vol. 62,pp. 189–193, 1967.

[39] S. Kreiner, “Analysis of multidimensional contingencey tablesby exact conditional tests,” Scandinavian Journal of Statistics,vol. 14, pp. 97–112, 1987.

[40] I. R. Bell, C. S. Miller, G. E. Schwartz, J. M. Peterson, andD. Amend, “Neuropsychiatric and somatic characteristics ofyoung adults with and without self-reported chemical odorintolerance and chemical sensitivity,” Archives of Environmen-tal Health, vol. 51, no. 1, pp. 9–21, 1996.

[41] J. B. Bjorner, S. Kreiner, J. E. Ware, M. T. Damsgaard, and P.Bech, “Differential item functioning in the Danish translationof the SF-36,” Journal of Clinical Epidemiology, vol. 51, no. 11,pp. 1189–1202, 1998.

[42] S. M. Caress and A. C. Stainamann, “National prevalenceof asthma and chemical hypersensitivity: an examination ofpotential overlap,” Journal of Occupational and EnvironmentalMedicine, vol. 47, no. 5, pp. 518–522, 2005.

[43] J. Elberling, A. Linneberg, A. Dirksen et al., “Mucosal symp-toms elicited by fragrance products in a population-basedsample in relation to atopy and bronchial hyper-reactivity,”Clinical and Experimental Allergy, vol. 35, no. 1, pp. 75–81,2005.

Submit your manuscripts athttp://www.hindawi.com

Stem CellsInternational

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014

Immunology ResearchHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinson’s Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttp://www.hindawi.com

evaluationofthequickenvironmentalexposureand...

Documents